The interaction of light and matter gives rise to a multitude of important and fascinating phenomena.
Computational studies of excited states are vital to further our basic understanding of these processes, and
design and optimise new processes for particular applications. However, the computational chemistry of
excited states gives rise to many challenging features, including differential static and dynamic correlation
effects, which can often be difficult to separate. Furthermore, regions of non-adiabatic coupling between
various potential energy surfaces are ubiquitous in photochemistry. Such regions where the Born-
Oppenheimer approximation breaks down are among the most difficult to treat.
The computational chemist must use a wide variety of methods to study photochemistry. However,
one important ‘tool’ in the computational arsenal is currently missing for general photochemical problems:
namely the ability to undertake systematically converging computations over all of the relevant regions of
the various (multi-state) potential energy surfaces. The Monte-Carlo Configuration Interaction (MC-CI)
method is ideal for this purpose, and has many desirable features, including automatic inclusion of strong
static correlation effects, and a balanced treatment of all states. Development of MC-CI methods, including
gradients and non-adiabatic couplings is proposed. This will give rise to the unprecedented ability to
benchmark a large variety of photochemical problems, across the entire potential energy surfaces, with
systematic accuracy. The method will be further extended by coupling within molecular mechanics in a
quantum mechanics / molecular mechanics (QM/MM) framework to study general excited state / open-shell
problems in complex environments.
The work will lead onto the applications research which spans the length scales of chemistry from
small molecules to large supramolecular systems. The above MC-CI method and other state-of-the-art
techniques will be applied to photochemical problems of enormous scientific interest. These include high
accuracy studies of inorganic photochemistry where the computational demands can be greatest, but also
where high-level electronic structure and dynamics simulation offers exceptional possibility to understand
complex molecular photochemistry. A practical area of photochemical research with a huge potential is
photodynamic therapy. Here light is used to destroy cancer tissue via the creation of the highly reactive
singlet molecular oxygen species. A deeper understanding of the many processes involved in this is required.
These include, single- vs multi-photon absorption, sensitizer internal conversion and intersystem crossing,
energy transfer processes with molecular oxygen, solvent effects, and aggregation effects. Detailed and
systematic studies of these fundamental aspects are proposed. The final applied area of study follows
naturally from this and is the supramolecular photochemistry of host-guest molecular sensors. Here advances
are required to allow a detailed understanding. These include the use of molecular dynamics simulation in
conjunction with QM/MM and statistical sampling.